Lev Core Wizard

Master orchestration for the Leviathan development environment.

Five-Fold Understanding

Before we begin, let me understand your quest:

1. Evolution - Fresh install or existing lev setup?
2. Impact - What are you trying to achieve? (dev environment, daemon management, debugging)
3. Relationships - Which project? (leviathan monorepo, external project, global setup)
4. Essence - The ONE thing blocking you right now?
5. Paradigm - Are you fighting lev or working with it?

Choose Your Path

What do you need help with?

1. Install & Setup (Legacy) - Historical setup/upgrade guidance → Load: references/install.md

2. Daemon Management - Start, stop, monitor poly daemons → Load: references/daemon.md

3. Debugging - Something broken? Troubleshoot common issues → Load: references/debug.md

4. Quick Reference - Just need a command? → See Quick Commands below

5. Something else? - Tell me your quest

Quick Commands

Command	Description
lev install	Legacy install path (prefer docs + runtime checks)
lev status	Show current system status
lev get <query>	Semantic search across codebase
lev check --log="msg"	Log progress checkpoint (lifecycle hook)
lev events tail	Tail lifecycle events
lev events query --type=X	Query events by type
lev events stats	Event statistics
lev daemon list	Show all registered daemons
lev daemon start <name>	Start a specific daemon
lev daemon stop <name>	Stop a specific daemon
lev daemon logs <name>	View daemon logs
bd ready	Show issues ready to work
bd stats	Project statistics

Architecture Overview

For full architecture details, run these bash commands against /lev or see: `/.agents/skills/lev/references/architecture-primer.md`

~/.config/lev/              # Global (user-level, XDG-compliant)
├── config.yaml             # Global settings
├── profiles/               # Exec profiles (planned)
└── runners/                # Runner protocol manifests (planned)

.lev/                       # Local (project-level)
├── config.yaml             # Project configuration (fractal: inline key → domain file → domain folder glob)
├── config/{domain}.yaml    # Per-domain config files
├── exec-profiles/          # Project exec profiles
└── pm/                     # Project management (handoffs, plans, intake)

core/                       # 20 modules (real layout)
├── domain/                 # Shared contracts: ProtocolAdapter, ExecTransport, Route, Target, Action
├── poly/                   # Protocol projector: registry.yaml, 5 surfaces (CLI/MCP/HTTP/gRPC/WS), codegen
├── exec/                   # Thin execution engine: createExec(dispatch), CLIAdapter, MCPAdapter
├── harness/                # Heavy runtime: TmuxHarness, AdapterRegistry (6 adapters), profile-loader
├── orchestration/          # Dispatch: A2A dispatcher, loop, DAG, stage policies
├── flowmind/               # Flow compilation + session management
├── graph/                  # Graph runtime: types, views, adapters, compositor
├── index/                  # Semantic search indexes
├── config/                 # Fractal config resolution
└── ...                     # + event-bus, event-machines, plugin-manager, etc.

Key Patterns

• ProtocolAdapter<TExternal, TInternal> — pure-function transforms between API conventions (domain)
• ExecTransport — domain defines interface, poly implements at boot (DI pattern)
• Route.parse('lev://module/fractal/item') — URI routing for everything (domain)
• A2A Job Dispatcher — AgentJob → JobExecutor → AgentJobResult (orchestration)
• TmuxHarness — owned spawn primitive: spawn/send/capture/kill on lev-<id> sessions (harness)
• Fractal Config — inline key → domain file → domain folder glob (NOT single config.yaml)
• DNA Contracts — 3 .dna.yaml files, 25+ constraints, gate vs aspirational enforcement

Protocols

Protocol	Purpose	Example
lev://	Internal routing	lev://poly/runners/claude
bin://	Binary handler dispatch	bin://ck
grpc://	gRPC service dispatch	grpc://leann:50052
http://	HTTP handler dispatch	http://deer-flow:8001
skill://	Load agent skills	skill://lev-core
bd://	Query beads issues	bd://lev-001b

External Tools (NOT our code)

• ntm — External Go tool for tmux agent spawning. We own TmuxHarness instead.
• bd/beads — Issue tracking (Rust CLI)
• ck — Code knowledge search (Rust CLI)

Related Skills

• bd - Issue tracking with dependency graphs
• lev-builder - Module placement decisions
• lev-cdo - Architecture-aware deliberation
• lev get - Semantic search across codebase (lev-find backend, legacy name)
• lev-index - Vector index management
• lev-learn - Event analysis and system improvement
• lev-align - Drift detection and remediation

Phase Navigation

Each reference follows the wizard pattern:

• 1-5 options presented based on context
• Combined choices welcome ("1,3")
• Pivot anytime ("actually, let's...")
• Back navigation available

When ready, tell me which path you'd like to take.

Technique Map

• Role definition - Clarifies operating scope and prevents ambiguous execution.
• Context enrichment - Captures required inputs before actions.
• Output structuring - Standardizes deliverables for consistent reuse.
• Step-by-step workflow - Reduces errors by making execution order explicit.
• Edge-case handling - Documents safe fallbacks when assumptions fail.

Technique Notes

These techniques improve reliability by making intent, inputs, outputs, and fallback paths explicit. Keep this section concise and additive so existing domain guidance remains primary.

Prompt Architect Overlay

Role Definition

You are the prompt-architect-enhanced specialist for lev-core, responsible for deterministic execution of this skill's guidance while preserving existing workflow and constraints.

Input Contract

• Required: clear user intent and relevant context for this skill.
• Preferred: repository/project constraints, existing artifacts, and success criteria.
• If context is missing, ask focused questions before proceeding.

Output Contract

• Provide structured, actionable outputs aligned to this skill's existing format.
• Include assumptions and next steps when appropriate.
• Preserve compatibility with existing sections and related skills.

Edge Cases & Fallbacks

• If prerequisites are missing, provide a minimal safe path and request missing inputs.
• If scope is ambiguous, narrow to the highest-confidence sub-task.
• If a requested action conflicts with existing constraints, explain and offer compliant alternatives.